SUMMARY Large-scale production of recombinant proteins in mammalian cells is the process for manufacturing protein biologics as medical therapies, for obtaining sufficient quantities of proteins for large-scale screens, and for creating protein bioconjugates. Many proteins must be made in mammalian cells (as opposed to bacteria) if the protein requires specific patterns of glycosylation, disulfide bond formation, or other modifications that can only be formed in mammalian cells. However, protein production in mammalian cells is much less efficient than protein production in bacterial cells. This is largely because the mRNA encoding the transgene is only a small fraction of the total cellular mRNA in mammalian cells. In contrast, in bacterial cells, the transgene mRNA can account for up to 50% of total cellular mRNA. In this application, we describe a strategy to increase transcript levels in mammalian cells by 50-100-fold compared to current levels. This quantum leap in transcript expression levels could fundamentally change protein production in mammalian cells. To do this, we will use the novel ?Tornado? technology for expressing transgenes as circular RNAs. These circular RNAs are exceptionally stable and can accumulate to exceptionally high levels compared to corresponding linear mRNAs. Chimerna scientists have generated key proof-of-concept data demonstrating the ability to express RNA circles containing inserts similar in size to protein-coding open reading frames. In order to develop a fundamentally new way to express proteins in mammalian cells at high levels, the specific aims of this proposal are: (1) To optimize protein expression from Tornado-derived circular RNAs. Circular RNAs can encode proteins if they contain an internal ribosome entry site (IRES). The goal of this aim is to characterize the optimal IRES, insert size, and cell line suitable for protein translation using Tornado-expressed circular RNA. We will systematically test each of these features and characterize protein output. (2) To compare protein output from Tornado-derived circular RNA and linear mRNAs. In this aim, we will compare protein output for cytosolic proteins and secreted proteins. We will directly compare linear to Tornado-encoded circular RNA and determine if the high-level circular RNA production achieved using the Tornado system results in increased protein production in cultured cells. Together, these experiments will allow us to test the idea that genetically encoded circular RNAs can serve as a new platform for high-level protein expression in mammalian cells. This expression system could have a major effect on biomedical research and protein manufacturing by reducing costs for protein manufacturing, increasing protein yields and simplifying protein expression.